arXiv subjects

Yu Wang

Publications and source records attributed to Yu Wang.

At least 325 records · Page 18Linked to original sources

arXiv2026

Physics Education under the Application of Artificial Intelligence: Bibliometric Analysis Based on Web of Science Core Library (2021-2025)

The rapid development of artificial intelligence technology is driving the transformation of physics education from traditional models to intelligent and data-driven approaches. To explore the evolution and cutting-edge hotspots in this field, this study conducted a systematic bibliometric analysis of 138 core literature published between 2021 and 2025 using VOSViewer and CiteSpace, based on the Web of Science Core Collection database. Research shows that the number of related publications will increase exponentially from 2023, with the United States, China, and Germany being the main research forces. The research focus has rapidly evolved from early machine learning assisted data analysis to the application of generative AI in teaching, the integration of physics information neural networks in computational physics courses, and the exploration of intelligent medical physics education. At present, this field is in the early stages of explosive growth and exhibits significant interdisciplinary characteristics. Future research should focus on building an adaptive learning ecosystem, reconstructing evaluation systems, and cultivating students' AI ethics and physical intuition.

physics.ed-ph↗

arXiv2026

MulCovFuzz: A Multi-Component Coverage-Guided Greybox Fuzzer for 5G Protocol Testing

As mobile networks transition to 5G infrastructure, ensuring robust security becomes more important due to the complex architecture and expanded attack surface. Traditional security testing approaches for 5G networks rely on black-box fuzzing techniques, which are limited by their inability to observe internal program state and coverage information. This paper presents MulCovFuzz, a novel coverage-guided greybox fuzzing tool for 5G network testing. Unlike existing tools that depend solely on system response, MulCovFuzz implements a multi-component coverage collection mechanism that dynamically monitors code coverage across different components of the 5G system architecture. Our approach introduces a novel testing paradigm that includes a scoring function combining coverage rewards with efficiency metrics to guide test case generation. We evaluate MulCovFuzz on open-source 5G implementation OpenAirInterface. Our experimental results demonstrate that MulCovFuzz significantly outperforms traditional fuzzing approaches, achieving a 5.85\% increase in branch coverage, 7.17\% increase in line coverage, and 16\% improvement in unique crash discovery during 24h fuzzing testing. MulCovFuzz uncovered three zero-day vulnerabilities, two of which were not identified by any other fuzzing technique. This work contributes to the advancement of security testing tools for next-generation mobile networks.

cs.CR↗

arXiv2026

APFuzz: Towards Automatic Greybox Protocol Fuzzing

Greybox protocol fuzzing is a random testing approach for stateful protocol implementations, where the input is protocol messages generated from mutations of seeds, and the search in the input space is driven by the feedback on coverage of both code and state. State model and message model are the core components of communication protocols, which also have significant impacts on protocol fuzzing. In this work, we propose APFuzz (Automatic greybox Protocol Fuzzer) with novel designs to increase the smartness of greybox protocol fuzzers from the perspectives of both the state model and the message model. On the one hand, APFuzz employs a two-stage process of static and dynamic analysis to automatically identify state variables, which are then used to infer an accurate state model during fuzzing. On the other hand, APFuzz introduces field-level mutation operations for binary protocols, leveraging message structure awareness enabled by Large Language Models. We conduct extensive experiments on a public protocol fuzzing benchmark, comparing APFuzz with the baseline fuzzer AFLNET as well as several state-of-the-art greybox protocol fuzzers.

cs.CR↗

arXiv2026

Human-Aligned MLLM Judges for Fine-Grained Image Editing Evaluation: A Benchmark, Framework, and Analysis

Evaluating image editing models remains challenging due to the coarse granularity and limited interpretability of traditional metrics, which often fail to capture aspects important to human perception and intent. Such metrics frequently reward visually plausible outputs while overlooking controllability, edit localization, and faithfulness to user instructions. In this work, we introduce a fine-grained Multimodal Large Language Model (MLLM)-as-a-Judge framework for image editing that decomposes common evaluation notions into twelve fine-grained interpretable factors spanning image preservation, edit quality, and instruction fidelity. Building on this formulation, we present a new human-validated benchmark that integrates human judgments, MLLM-based evaluations, model outputs, and traditional metrics across diverse image editing tasks. Through extensive human studies, we show that the proposed MLLM judges align closely with human evaluations at a fine granularity, supporting their use as reliable and scalable evaluators. We further demonstrate that traditional image editing metrics are often poor proxies for these factors, failing to distinguish over-edited or semantically imprecise outputs, whereas our judges provide more intuitive and informative assessments in both offline and online settings. Together, this work introduces a benchmark, a principled factorization, and empirical evidence positioning fine-grained MLLM judges as a practical foundation for studying, comparing, and improving image editing approaches.

cs.CV↗

arXiv2026

SaVe-TAG: LLM-based Interpolation for Long-Tailed Text-Attributed Graphs

Real-world graph data often follows long-tailed distributions, making it difficult for Graph Neural Networks (GNNs) to generalize well across both head and tail classes. Recent advances in Vicinal Risk Minimization (VRM) have shown promise in mitigating class imbalance with numeric interpolation; however, existing approaches largely rely on embedding-space arithmetic, which fails to capture the rich semantics inherent in text-attributed graphs. In this work, we propose our method, SaVe-TAG (Semantic-aware Vicinal Risk Minimization for Long-Tailed Text-Attributed Graphs), a novel VRM framework that leverages Large Language Models (LLMs) to perform text-level interpolation, generating on-manifold, boundary-enriching synthetic samples for minority classes. To mitigate the risk of noisy generation, we introduce a confidence-based edge assignment mechanism that uses graph topology as a natural filter to ensure structural consistency. We provide theoretical justification for our method and conduct extensive experiments on benchmark datasets, showing that our approach consistently outperforms both numeric interpolation and prior long-tailed node classification baselines. Our results highlight the importance of integrating semantic and structural signals for balanced and effective learning on text-attributed graphs. The source code is publicly available at: https://github.com/LWang-Laura/SaVe-TAG.

cs.AI↗

arXiv2026

Translate Policy to Language: Flow Matching Generated Rewards for LLM Explanations

As humans increasingly share environments with diverse agents powered by RL, LLMs, and beyond, the ability to explain agent policies in natural language is vital for reliable coexistence. We introduce a general-purpose framework that trains explanation-generating LLMs via reinforcement learning from AI feedback, with distributional rewards generated by generative continuous normalizing flows (CNFs). CNFs capture the pluralistic and probabilistic nature of human judgments about explanations. Moreover, under mild assumptions, CNFs provably bound deviations from true human reward distributions when trained on noisy proxy rewards from LLMs. We design a specialized CNF architecture that selectively attends to linguistic cues in the decision context and explanations when generating rewards. Human and LLM evaluators find that our method delivers explanations that enable more accurate predictions of true agent decisions, exhibit greater logical soundness and actionability, and impose lower cognitive load than explanations trained with proxy LLM rewards or state-of-the-art RLHF and RLAIF baselines.

cs.CL↗

arXiv2026

MARSHAL: Incentivizing Multi-Agent Reasoning via Self-Play with Strategic LLMs

Developing Large Language Models (LLMs) to cooperate and compete effectively within multi-agent systems (MASs) is a critical step towards more advanced intelligence. While reinforcement learning (RL) has proven effective for enhancing reasoning in single-agent tasks, its extension to multi-turn, multi-agent scenarios remains underexplored due to the challenges of long-horizon credit assignment and agent-specific advantage estimation. To address these challenges, we introduce MARSHAL, an end-to-end RL framework that incentivizes Multi-Agent Reasoning through Self-play witH strAtegic LLMs in both cooperative and competitive games. MARSHAL features a turn-level advantage estimator that aligns learning signals with each interaction for credit assignment, and an agent-specific advantage normalization to stabilize multi-agent training. By learning with self-play across cooperative and competitive games, MARSHAL agents trained from Qwen3-4B develop strong strategic abilities, with up to 28.7% performance improvements in held-out games. More importantly, the capability acquired through self-play generalizes beyond games, yielding consistent performance gains of MASs in reasoning benchmarks. When integrated into leading MASs, our MARSHAL agent achieves significant zero-shot performance gains of up to 10.0% on AIME, 7.6% on GPQA-Diamond, and 3.5% on average across all benchmarks. These results establish self-play in strategic games as a powerful approach for developing generalizable multi-agent reasoning capabilities in LLMs.

cs.AI↗

arXiv2026

LCIP: Loss-Controlled Inverse Projection of High-Dimensional Image Data

Projections (or dimensionality reduction) methods $P$ aim to map high-dimensional data to typically 2D scatterplots for visual exploration. Inverse projection methods $P^{-1}$ aim to map this 2D space to the data space to support tasks such as data augmentation, classifier analysis, and data imputation. Current $P^{-1}$ methods suffer from a fundamental limitation -- they can only generate a fixed surface-like structure in data space, which poorly covers the richness of this space. We address this by a new method that can `sweep' the data space under user control. Our method works generically for any $P$ technique and dataset, is controlled by two intuitive user-set parameters, and is simple to implement. We demonstrate it by an extensive application involving image manipulation for style transfer.

cs.HC↗

arXiv2026

VocalNet-MDM: Accelerating Streaming Speech LLM via Self-Distilled Masked Diffusion Modeling

Recent Speech Large Language Models~(LLMs) have achieved impressive capabilities in end-to-end speech interaction. However, the prevailing autoregressive paradigm imposes strict serial constraints, limiting generation efficiency and introducing exposure bias. In this paper, we investigate Masked Diffusion Modeling~(MDM) as a non-autoregressive paradigm for speech LLMs and introduce VocalNet-MDM. To adapt MDM for streaming speech interaction, we address two critical challenges: training-inference mismatch and iterative overhead. We propose Hierarchical Block-wise Masking to align training objectives with the progressive masked states encountered during block diffusion decoding, and Iterative Self-Distillation to compress multi-step refinement into fewer steps for low-latency inference. Trained on a limited scale of only 6K hours of speech data, VocalNet-MDM achieves a 3.7$\times$--10$\times$ decoding speedup and reduces first-chunk latency by 34\% compared to AR baselines. It maintains competitive recognition accuracy while achieving state-of-the-art text quality and speech naturalness, demonstrating that MDM is a promising and scalable alternative for low-latency, efficient speech LLMs.

cs.CL↗

arXiv2026

Lithium niobate quadratic integrated nonlinear photonics: enabling ultra-wide bandwidth and ultrafast photonic engines

Integrated photonic coherent light sources capable of generating emission with broad spectral coverage and ultrashort pulse durations are critical for both fundamental science and emerging technologies. In this Perspective, we start by discussing emerging quantum and classical photonic applications from the standpoint of operating wavelength and timescale, highlighting the technological gaps that persist in current integrated photonic light sources. Next, we introduce the unique properties of lithium niobate-based integrated quadratic nonlinear photonics, and discuss several promising strategies that exploit this platform to realize wavelength-tunable continuous wave light sources and broadband, ultra-short light pulse generation. We also assessed their advantages and limitations while discussing potential solutions. Finally, we outline future prospects and challenges that need to be addressed, aiming at inspiring continued research and innovation in this rapidly evolving field.

physics.optics↗

arXiv2026

LVK S241125n: Massive Binary Black Hole Merger Produces Gamma Ray Burst in Active Galactic Nucleus Disk

Recently, the gravitational-wave (GW) event S241125n, detected by LIGO/Virgo/KAGRA (LVK), has been reported to coincide with a candidate detected by Swift-BAT/GUANO and an X-ray candidate found by FXT onboard of Einstein Probe (EP) and confirmed by Swift-XRT. We estimate that the joint false alarm rate (FAR) for the three candidates is 1 / 30 yr and that the corresponding false alarm probability (FAP) is $\mathrm{FAP}_{\rm triple} = 0.037$ ($1.8 σ$). The coincidence between the GW and GRB could be an interesting test of their origin and open attractive opportunities for multi-messenger observations, if they are actually associated. Motivated by this, we propose a theoretical model in which a binary black hole (BBH) merger occurs within an active galactic nucleus (AGN) disk. The typically massive and significantly kicked merger remnant accretes disk material at hyper-Eddington rates, and the resulting jet could lead to the GRB associated with the GW event. As the jet interacts with the gas in the AGN disk, the shock breakout produces a Comptonized spectrum, consistent with an unusually soft photon index of the GRB prompt emission observed by Swift-BAT following LVK S241125n. Meanwhile, strong absorption and dust extinction of the afterglow by the high column density typical of AGN disks could explain the unusually hard spectrum observed in the X-ray band by EP, as well as the non-detection of an optical counterpart. Our model is predictive, and we highlight the importance of further constraining the orbital eccentricity of the merger and conducting deep-field observations of the host galaxy to test our explanation.

astro-ph.HE↗

arXiv2026

Learning to Self-Verify Makes Language Models Better Reasoners

Recent large language models (LLMs) achieve strong performance in generating promising reasoning paths for complex tasks. However, despite powerful generation ability, LLMs remain weak at verifying their own answers, revealing a persistent capability asymmetry between generation and self-verification. In this work, we conduct an in-depth investigation of this asymmetry throughout training evolution and show that, even on the same task, improving generation does not lead to corresponding improvements in self-verification. Interestingly, we find that the reverse direction of this asymmetry behaves differently: learning to self-verify can effectively improve generation performance, achieving accuracy comparable to standard generation training while yielding more efficient and effective reasoning traces. Building on this observation, we further explore integrating self-verification into generation training by formulating a multi-task reinforcement learning framework, where generation and self-verification are optimized as two independent but complementary objectives. Extensive experiments across benchmarks and models demonstrate performance gains over generation-only training in both generation and verification capabilities.

cs.CL↗

arXiv2026

CineScene: Implicit 3D as Effective Scene Representation for Cinematic Video Generation

Cinematic video production requires control over scene-subject composition and camera movement, but live-action shooting remains costly due to the need for constructing physical sets. To address this, we introduce the task of cinematic video generation with decoupled scene context: given multiple images of a static environment, the goal is to synthesize high-quality videos featuring dynamic subject while preserving the underlying scene consistency and following a user-specified camera trajectory. We present CineScene, a framework that leverages implicit 3D-aware scene representation for cinematic video generation. Our key innovation is a novel context conditioning mechanism that injects 3D-aware features in an implicit way: By encoding scene images into visual representations through VGGT, CineScene injects spatial priors into a pretrained text-to-video generation model by additional context concatenation, enabling camera-controlled video synthesis with consistent scenes and dynamic subjects. To further enhance the model's robustness, we introduce a simple yet effective random-shuffling strategy for the input scene images during training. To address the lack of training data, we construct a scene-decoupled dataset with Unreal Engine 5, containing paired videos of scenes with and without dynamic subjects, panoramic images representing the underlying static scene, along with their camera trajectories. Experiments show that CineScene achieves state-of-the-art performance in scene-consistent cinematic video generation, handling large camera movements and demonstrating generalization across diverse environments.

cs.CV↗

arXiv2026

Balanced Anomaly-guided Ego-graph Diffusion Model for Inductive Graph Anomaly Detection

Graph anomaly detection (GAD) is crucial in applications like fraud detection and cybersecurity. Despite recent advancements using graph neural networks (GNNs), two major challenges persist. At the model level, most methods adopt a transductive learning paradigm, which assumes static graph structures, making them unsuitable for dynamic, evolving networks. At the data level, the extreme class imbalance, where anomalous nodes are rare, leads to biased models that fail to generalize to unseen anomalies. These challenges are interdependent: static transductive frameworks limit effective data augmentation, while imbalance exacerbates model distortion in inductive learning settings. To address these challenges, we propose a novel data-centric framework that integrates dynamic graph modeling with balanced anomaly synthesis. Our framework features: (1) a discrete ego-graph diffusion model, which captures the local topology of anomalies to generate ego-graphs aligned with anomalous structural distribution, and (2) a curriculum anomaly augmentation mechanism, which dynamically adjusts synthetic data generation during training, focusing on underrepresented anomaly patterns to improve detection and generalization. Experiments on five datasets demonstrate that the effectiveness of our framework.

cs.LG↗

arXiv2026

Explainable Pathomics Feature Visualization via Correlation-aware Conditional Feature Editing

Pathomics is a recent approach that offers rich quantitative features beyond what black-box deep learning can provide, supporting more reproducible and explainable biomarkers in digital pathology. However, many derived features (e.g., "second-order moment") remain difficult to interpret, especially across different clinical contexts, which limits their practical adoption. Conditional diffusion models show promise for explainability through feature editing, but they typically assume feature independence**--**an assumption violated by intrinsically correlated pathomics features. Consequently, editing one feature while fixing others can push the model off the biological manifold and produce unrealistic artifacts. To address this, we propose a Manifold-Aware Diffusion (MAD) framework for controllable and biologically plausible cell nuclei editing. Unlike existing approaches, our method regularizes feature trajectories within a disentangled latent space learned by a variational auto-encoder (VAE). This ensures that manipulating a target feature automatically adjusts correlated attributes to remain within the learned distribution of real cells. These optimized features then guide a conditional diffusion model to synthesize high-fidelity images. Experiments demonstrate that our approach is able to navigate the manifold of pathomics features when editing those features. The proposed method outperforms baseline methods in conditional feature editing while preserving structural coherence.

cs.CV↗

arXiv2026

LLM-Driven Transient Stability Assessment: From Automated Simulation to Neural Architecture Design

This paper presents an LLM-driven, end-to-end workflow that addresses the lack of automation and intelligence in power system transient stability assessment (TSA). The proposed agentic framework integrates large language models (LLMs) with a professional simulator (ANDES) to automatically generate and filter disturbance scenarios from natural language, and employs an LLM-driven Neural Network Design (LLM-NND) pipeline to autonomously design and optimize TSA models through performance-guided, closed-loop feedback. On the IEEE 39-bus system, the LLM-NND models achieve 93.71% test accuracy on four-class TSA with only 4.78M parameters, while maintaining real-time inference latency (less than 0.95 ms per sample). Compared with a manually designed DenseNet (25.9M parameters, 80.05% accuracy), the proposed approach jointly improves accuracy and efficiency. Ablation studies confirm that the synergy among domain-grounded retrieval, reasoning augmentation, and feedback mechanisms is essential for robust automation. The results demonstrate that LLM agents can reliably accelerate TSA research from scenario generation and data acquisition to model design and interpretation, offering a scalable paradigm that is readily extensible to other power system tasks such as optimal power flow, fault analysis, and market operations.

eess.SY↗

arXiv2026

Growth of High-Redshift Quasars from Fermion Dark Matter Seeds

Quasars hosting $\gtrsim 10^{9}\,M_\odot$ black holes at $z>6$ challenge growth scenarios that start from light seeds and assume accretion within already formed galaxies. Motivated by the James Webb Space Telescope (JWST) discovery of Little Red Dots (LRDs), which suggests that $\sim 10^{6}\,M_\odot$ black holes can be active in compact, pre-galactic environments, we revisit early black hole growth with a minimal cosmology-based framework. We model the accretion history as the smaller of the Bondi inflow rate and the Eddington-limited rate, where the Bondi rate is set by the supply of overdense primordial gas whose density declines with cosmic expansion, and the Eddington rate captures regulation by radiative feedback. By fitting the observed masses and luminosities of J0313--1806 ($z=7.64$) and J0100+2802 ($z=6.30$) with Bayesian inference, we infer initial conditions that favor massive seed black holes with initial mass $M_0 \sim 10^{6}\,M_\odot$, formed at $z\sim20{-}30$ in environments with baryonic overdensity factors $f_ρ\gtrsim 50$ relative to the cosmic mean. The resulting growth histories include a prolonged supply-limited stage, and they reproduce the observed quasar masses without requiring sustained Eddington accretion or any super-Eddington episodes. The inferred seed mass scale is consistent with black holes produced by the collapse of quantum-degenerate fermion dark matter cores, providing a physically defined pathway to massive seeds at the redshifts implied by LRD phenomenology.

astro-ph.GA↗

arXiv2026

Reliable and Responsible Foundation Models: A Comprehensive Survey

Foundation models, including Large Language Models (LLMs), Multimodal Large Language Models (MLLMs), Image Generative Models (i.e, Text-to-Image Models and Image-Editing Models), and Video Generative Models, have become essential tools with broad applications across various domains such as law, medicine, education, finance, science, and beyond. As these models see increasing real-world deployment, ensuring their reliability and responsibility has become critical for academia, industry, and government. This survey addresses the reliable and responsible development of foundation models. We explore critical issues, including bias and fairness, security and privacy, uncertainty, explainability, and distribution shift. Our research also covers model limitations, such as hallucinations, as well as methods like alignment and Artificial Intelligence-Generated Content (AIGC) detection. For each area, we review the current state of the field and outline concrete future research directions. Additionally, we discuss the intersections between these areas, highlighting their connections and shared challenges. We hope our survey fosters the development of foundation models that are not only powerful but also ethical, trustworthy, reliable, and socially responsible.

cs.LG↗